First Align, then Predict: Understanding the Cross-Lingual Ability of Multilingual BERT - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Preprints, Working Papers, ... Year :

First Align, then Predict: Understanding the Cross-Lingual Ability of Multilingual BERT

Abstract

Multilingual pretrained language models have demonstrated remarkable zero-shot cross-lingual transfer capabilities. Such transfer emerges by fine-tuning on a task of interest in one language and evaluating on a distinct language, not seen during the fine-tuning. Despite promising results, we still lack a proper understanding of the source of this transfer. Using a novel layer ablation technique and analyses of the model's internal representations, we show that multilingual BERT, a popular multilingual language model, can be viewed as the stacking of two sub-networks: a multilingual encoder followed by a task-specific language-agnostic predictor. While the encoder is crucial for cross-lingual transfer and remains mostly unchanged during fine-tuning, the task predictor has little importance on the transfer and can be reinitialized during fine-tuning. We present extensive experiments with three distinct tasks, seventeen typologically diverse languages and multiple domains to support our hypothesis.

Dates and versions

hal-03161685 , version 1 (08-03-2021)

Identifiers

Cite

Benjamin Muller, Yanai Elazar, Benoît Sagot, Djamé Seddah. First Align, then Predict: Understanding the Cross-Lingual Ability of Multilingual BERT. 2021. ⟨hal-03161685⟩
30 View
0 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More